Low Cost Hardware Back Propagation Algorithm
Abstract
The first successful implementation of Artificial Neural Networks (ANNs) was published a little over a decade ago. It is time to review the progress that has been made in this research area. This paper provides taxonomy for classifying Field Programmable Gate Arrays (FPGAs) implementation of ANNs. Different implementation techniques and design issues are discussed, such as obtaining a suitable activation function and numerical truncation technique trade-off, the improvement of the learning algorithm to reduce the cost of neuron and in result the total cost and the total speed of the complete ANN. Finally, the implementation of a complete very fast circuit for the pattern of English Digit Numbers NN has four layers of 70 nodes (neurons) on single chip using Xilinx FPGA technique is given.
The main goal of this paper is how to achieve the suitable activation function and weights for this network that gives minimum hardware cost when all stages of this ANN algorithm is implemented on FPGA.
Downloads
References
[2] J. G. ELDREDGE, “FPGA DENSITY ENHANCEMENT OF A NEURAL NETWORK THROUGH RUN-TIME RECONFIGURATION”, MASTER’S THESIS, DEPARTMENT OF ELECTRICAL AND COMPUTER ENGINEERING, BRIGHAM YOUNG UNIVERSITY, MAY 1994.
[3] D.E RUMELHART, J.L MCCLELLAND AND PDP RESEARCH GROUP, PARALLEL DISTRUBUTED PROCESSING: EXPLORATIONS IN THE MICROSTRUCTURE OF COGNITION, VOLUME 1: FOUNDATIONS, MIT PRESS, CAMBRIDGE, MASSACHUSETTS, 1986.
[4] A. DEHON, “THE DENSITY ADVANTAGE OF CONFIGURABLE COMPUTING”, IEEE COMPUTER, VOL. 33, N. 5, PP. 41–49, APRIL 2000.
[5] HOLT, J.L., T.E. BAKER. BACK PROPAGATION SIMULATIONS USING LIMITED PRECISION CALCULATIONS, IN PROCEEDINGS OF INTERNATIONAL JOINT CONFERENCE ON NEURAL NETWORKS. 1991. PP 121-126 VOL. 2.
[6] E. K. KNIGHT, “ARTIFICIAL INTELLINGENCE” MCGRAW-HILL NEW YORK, SECOND EDITION, 1991.
[7] XILINX.COM, “VIRTEXTM2.5V FPGAS”, DATA SHEET, WWW.XILINX.COM, MAY 1999.
[8] WOLF, D.F., ROMERO, R. A. F., MARQUES, E. USING EMBEDDED PROCESSORS IN HARDWARE MODELS OF ARTIFICIAL NEURAL NETWORKS. IN PROCEEDINGS OF SBAI - SIMPَSIO BRASILEIRO DE AUTOMAO INTELIGENTE. 2001. PP 78-83.
Downloads
Published
Issue
Section
License
Copyright: Open Access authors retain the copyrights of their papers, and all open access articles are distributed under the terms of the Creative Commons Attribution License, which permits unrestricted use, distribution and reproduction in any medium, provided that the original work is properly cited. The use of general descriptive names, trade names, trademarks, and so forth in this publication, even if not specifically identified, does not imply that these names are not protected by the relevant laws and regulations. While the advice and information in this journal are believed to be true and accurate on the date of its going to press, neither the authors, the editors, nor the publisher can accept any legal responsibility for any errors or omissions that may be made. The publisher makes no warranty, express or implied, with respect to the material contained herein.
